RMA Week 7: Descriptive Statistics I

if a variable has high testability, it is…

easy to test

what’s a measurement

a way to describe real life factors by numbers

what are the 4 types of measurement (scales)

1. nominal

2. ordinal

3. interval

4. ratio

how do the scales differ?

• the properties of the numbers e.g. 2 v 4

• the properties of what is being measured e.g. does a score of 4 = twice 2

describe features and an example of nominal scales

• numbers are merely labels

• no relationship between size of number and attribute measured

• e.g. numbers of buildings on university map

describe features and an example of ordinal scales

• order of size of numbers = order of size of attribute measured

• data are put in order, but it’s only a relative ranking - there is no relationship between the numbers - distance between scores vary

• e.g. IQ scores

describe features and an example of interval scales

• equal intervals on the scale = equal intervals in the property measured

• measure of magnitude

• zero point on the scale is arbitrary - you can get negative values

• e.g. degrees celsius - difference between 10 and 20 is the same as 20 and 30, but 20 is not as twice as warm as 10.

• zero is not a meaningful point e.g. -10 exists.

describe features and an example of ratio scales

• it’s an interval scale except the number zero denotes absolute absence!

• e.g. reaction time - cannot have a negative

the mean

sum of numbers/number of numbers

median

ranked in ascending order, choose middle number. if number of numbers even, the median is the mean of the two central scores.

when to use mean v median

mean = use if all values important and valid, and if data is interval or ratio

median = use if outliers (as its less affected by them) and if ordinal data

mode

most frequently occuring score

when to use mode

if data on nominal scale

limitations of the mode

some data doesn’t have a mode or can be bimodal. the mode can be atypical.